Minimally invasive cardiac procedures using endovascular access under x-ray fluoroscopic guidance have enabled treatment of patients that previously would have required open-heart surgery. Such minimally invasive procedures are preferred because of their lower risk, faster recovery and lower cost. One of the fastest growing procedures is ablation for treatment of atrial fibrillation - a problem eventually suffered by 20% of the population and a significant contributor to stroke, patient morbidity and mortality. While promising, procedure success rates are still only ~50% for the first intervention, and even with subsequent interventions, final outcome success hovers at ~70%. It is hypothesized that a major factor contributing to the limited success rate is the lack of precise knowledge about the exact number and location of the created lesions. We propose to leverage and expand upon our recent development of ECG gated cardiac C-arm CT to provide three-dimensional imaging during the interventional procedure. By generating 3D images utilizing the x-ray system in the interventional suite, the error prone and time-consuming spatio-temporal registration with prior CT or MRI exams will be avoided. Detailed three- dimensional images reflecting the current state of the patient will be combined with electroanatomical maps to optimally assist the interventionalist. We aim to also reduce the radiation dose to both patients and staff by shortening the time needed for fluoroscopic guidance and by incorporating multi-dimensional noise reduction algorithms. We will develop image quality improvements by addressing under-sampling, acquisition timing and temporal definition, and by developing hardware and software to improve low-contrast performance, registration accuracy and visualization. One of our goals is to demonstrate the ability to depict the left atrium and pulmonary veins to provide guidance accuracy of better then 2 mm, and to reduce the time required for registration by 50% (~20 minutes). With these developments our ultimate goal is to accomplish faster and more accurate catheter based interventions with an equivalent success rate as surgical procedures (80-90%). The significance of this project is further amplified by the fact that the technology developed could eventually impact many other current minimally invasive procedures, and also future procedures as they become available, such as endocardial injection of stem cells for treatment of myocardial infarction. Project Narrative The project aims to provide high-quality 3D CT images of the beating heart in the interventional suite during an interventional procedure. We believe that providing images that reflect the true status of the patient at the time of intervention, when combined with device-tracking hardware, will provide significant improvements in accuracy and reduction in procedure times for complicated procedures such as radiofrequency ablation of atrial fibrillation. We are also developing new image processing and reconstruction approaches that may permit 3D visualization of radiofrequency lesions in the heart wall that would allow intra-procedural monitoring of lesion placement. Access to such images, during an interventional procedure, could also enable new, directed treatments for myocardial infarction by allowing accurate guidance of therapy (cell, drug etc.) directly to the affected tissue.